Sphingosine-1-phosphate receptor 1 mediates elevated IL-6 signaling to promote chronic inflammation and multitissue damage in sickle cell disease

Abstract

Sphingosine-1-phosphate (S1P) is a biolipid involved in chronic inflammation in several inflammatory disorders. Recent studies revealed that elevated S1P contributes to sickling in sickle cell disease (SCD), a devastating hemolytic, genetic disorder associated with severe chronic inflammation and tissue damage. We evaluated the effect of elevated S1P in chronic inflammation and tissue damage in SCD and underlying mechanisms. First, we demonstrated that interfering with S1P receptor signaling by FTY720, a U.S. Food and Drug Administration–approved drug, significantly reduced systemic, local inflammation and tissue damage without antisickling effects. These findings led us to discover that S1P receptor activation leads to substantial elevated local and systemic IL-6 levels in SCD mice. Genetic deletion of IL-6 in SCD mice significantly reduced local and systemic inflammation, tissue damage, and kidney dysfunction. At the cellular level, we determined that elevated IL-6 is a key cytokine functioning downstream of elevated S1P, which contributes to increased S1P receptor 1 (S1pr1) gene expression in the macrophages of several tissues in SCD mice. Mechanistically, we revealed that S1P-S1PR1 signaling reciprocally up-regulated IL-6 gene expression in primary mouse macrophages in a JAK2-dependent manner. Altogether, we revealed that elevated S1P, coupled with macrophage S1PR1 reciprocally inducing IL-6 expression, is a key signaling network functioning as a malicious, positive, feed-forward loop to sustain inflammation and promote tissue damage in SCD. Our findings immediately highlight novel therapeutic possibilities.—Zhao, S., Adebiyi, M. G., Zhang, Y., Couturier, J. P., Fan, X., Zhang, H., Kellems, R. E., Lewis, D. E., Xia, Y. Sphingosine-1-phosphate receptor 1 mediates elevated IL-6 signaling to promote chronic inflammation and multitissue damage in sickle cell disease.

Sphingosine-1-phosphate (S1P) is an important signaling sphingolipid (1) involved in physiologic and pathophysiologic processes, including cell proliferation (2), endothelial integrity (3), angiogenesis, and inflammation (4, 5), by activating its 5 receptors (S1PR1–5). Moreover, several studies determined that S1P functions intracellularly by interacting with specific molecules, including TRAF2, HDAC1, and BACE1, to regulate ubiquitination (6), histone deacetylation, and BACE1 activity (7, 8). S1P is produced intracellularly from sphingosine by 2 sphingosine kinases, SPHK1 and SPHK2, and can be dephosphorylated to sphingosine by S1P phosphatases 1 and 2 (SGPP1 and SGPP2) (9). S1P is also irreversibly degraded to fatty aldehydes and phosphoethanolamine by S1P lyase (9). In most cells, intracellular S1P is usually maintained at a low level because of high S1P phosphatases and/or S1P lyase activity. However, because of relatively high SPHK1 activity and lack of S1P-degrading enzymes, erythrocytes store high levels of S1P and are considered the major cell type for supplying S1P to plasma by a regulated release process that is poorly understood (9–11). Thus, the concentration of S1P in the plasma is much greater than it is in tissues. Early studies showed that this gradient is extremely important for lymphocyte (LY) trafficking and egress from spleen, thymus, and lymph nodes via S1PR1 activation (12). Additional studies demonstrated that S1P is a key chemoattractant to induce myeloid progenitor cells to egress from bone marrow to peripheral blood (13). Besides inducing egress of LYs and myeloid progenitor cells, S1P is a potent inflammatory mediator activating its specific receptors to induce chronic inflammation by promoting monocyte survival (14, 15), triggering neutrophil activation (16), and sustaining persistently elevated IL-6 production (5, 17). Thus, plasma S1P signaling is important for chronic and prolonged inflammatory responses, which include immune cell trafficking and egress, activation of immune cells, and persistently up-regulating inflammatory cytokine secretion.

Sickle cell disease (SCD) is the most prevalent, inherited blood disorder, affecting millions worldwide. It is caused by a missense mutation in the β-globin gene (18). This mutation results in the synthesis of an abnormal hemoglobin, termed hemoglobin S. Under hypoxic conditions, deoxygenated hemoglobin S forms polymers that cause normal donut-shaped erythrocytes to become sickle shaped. Sickling leads to increased intravascular hemolysis and profound inflammation. Without interference, inflammation will cause vasoocclusion, multitissue damage, and early death (19). Although sickling is the central pathogenic process of the disease, inflammation is a severe complication seen in both humans and mice with SCD (20, 21). However, the molecular basis responsible for profound inflammation in SCD is not fully understood. Thus, identification of specific factors and signaling pathways involved in chronic inflammation and disease progression of SCD are extremely important and will likely highlight potential new, specific therapeutic targets to reduce prolonged inflammation and slow disease progression.

A recent metabolomic study revealed that erythrocyte-increased SPHK1 activity mediates production of elevated S1P, which, in turn, functions intracellularly inducing sickling and disease progression (22). Sickle cells are fragile and prone to lysis when passing through capillary vessels, resulting in intravascular hemolysis. Because erythrocytes are the largest reservoir of S1P (9, 10, 23), the intravascular hemolysis associated with SCD unleashes massive amounts of this pleiotropic signaling molecule into the plasma. Supporting that notion, recent studies provide evidence that plasma S1P levels are significantly elevated in SCD mice and patients with SCD (22, 24). However, the functional consequences of elevated circulating S1P in the pathophysiology of SCD remain unidentified. Given the critical role of chronic inflammation in multitissue damage and disease progression in SCD, we sought to determine the role of elevated extracellular S1P in chronic inflammation and tissue damage in SCD and underlying mechanisms.

MATERIALS AND METHODS

Reagents

FTY720 was purchased from Cayman Chemicals (Ann Arbor, MI, USA). W146, JTE013, TY52156, CYM50358, and AG490 were R&D Systems brand (Bio-Techne, Minneapolis, MN, USA). Cell-culture medium, antibiotics, and fetal bovine serum were purchased from Thermo Fisher Scientific (Waltham, MA, USA).

Mice

Wild-type (WT), 8–10-wk-old mice were purchased from Envigo (East Millstone, NJ, USA). SCD mice (Berkeley SCD mice) were originally purchased from The Jackson Laboratory (Bar Harbor, ME, USA) and bred in the animal facility at The University of Texas at Houston, Health Science Center (Houston, TX, USA). All mice were housed and treated in accordance with protocols approved by the Center for Laboratory Animal Medicine and Care at The University of Texas at Houston, Health Science Center.

Generation of SCD/IL-6^−/− mice

SCD/Il6^−/− mice were generated by mating SCD mice with Il6^−/− mice, as previously described (25). Specifically, 2 SCD heterozygote, female mice were bred with 1 Il6^−/−, male mouse to generate hemoglobin α (Hbα), hemoglobin β (Hbβ), and IL-6 heterozygous (Hbα^+/−/Hbβ^+/−/Il6^+/−) mice. Then, 2 Hbα^+/−/Hbβ^+/−/Il6^+/− heterozygote, female mice were bred with 1 heterozygote, male mouse to generate the offspring of Hbα^−/−/Hbβ^+/−/Il6^−/− female and male mice. Finally, Hbα^−/−/Hbβ^+/−/Il6^−/− female and male mice were chosen for breeding to generate mice with Hbα^−/−/Hbβ^−/−/Il6^−/− (SCD/Il6^−/−). Genotyping methods included mouse genomic DNA, which was isolated from the clipped mouse tail. Hbα, Hbβ, and IL-6 genes were detected, respectively, by PCR as described by The Jackson Laboratory. Primers for Hbα included IMR1137, 5′-AGTGGGCAGCTTCTAACTATGC-3′; IMR1138, 5′-GTCCCAGCGCATACCTTG-3′; IMR1139, 5′-ATAGATGGGTAGCCATTTAGATTCC-3′; and IMR1140, 5′-CCGGGTTATAATTACCTCAGGTC-3′. The PCR program for Hbα was 94°C for 3 min, 94°C for 30 s, 52°C for 45 s, 72°C for 45 s, repeated steps 2–4 for 35 cycles, 72°C for 2 min, and a 10°C hold. WT mouse had a 280–base pair band, and the mutated mouse had a 461–base pair band. Primers for Hbβ included IMR1141, 5′-TTGGTGGTCTTAAAACTTTTGTGG-3′; IMR1142, 5′-ACTGGCACAGAGCTTGTTATG-3′; IMR1143, 5′-AGATGTTTTTTTCACATTCTTGAGC-3′; and IMR1144, 5′-AATGCCTGCTCTTTACTGAAGG-3′. The PCR program was the same as that for Hbα. The WT mouse had a 291–base pair band, and the mutated mouse had a 398–base pair band. Primers for IL-6 knockout included IL-6 forward, 5′-TTCCATCCAGTTGCCTTCTTGG-3′; IL-6 reverse, 5′-TTCTCATTTCCACGATTTCCCAG-3′; and IL-6 Neo, 5′-CCGGAGAACCTGCGTGCAATCC-3′. The PCR program was 94°C for 5 min, 84°C for 10 min, 94°C for 1 min, 60°C for 30 s, 72°C for 2 min, repeated steps 3–5 for 40 cycles, 72°C for 5 min, and a 4°C hold. WT mouse had a 174–base pair band, and the mutated mouse had a 380–base pair band. Genotypes were summarized and added to Supplemental Fig. 2.

Hematologic analysis in SCD mice

Peripherial blood from mice was collected in EDTA-anticoagulant tubes. Complete blood cell count (CBC) was analyzed with a ProCYTE DX hematology analyzer (IDEXX Laboratories, Westbrook, ME, USA).

ELISA detection of IL-2, -12, -17A, and -6 in mouse circulation

Blood was drawn via cardiac puncture and collected in EDTA-anticoagulant tubes. Blood was centrifuged at 2400 g for 5 min to collect serum. IL-2, IL-12, and IL-17A were detected with a Multi-Analyte ELISA Array Kit (MEM-004A; Qiagen, Hilden, Germany). Plasma and tissue levels of IL-6 were determined with a single-cytokine detection kit (EZMIL-6; MilliporeSigma, Billerica, MA, USA).

Measurement of life span of erythrocytes

Erythrocytes life span was detected by in vivo injection of N-hydroxysuccinimide biotin (50 mg/kg) into retro-orbital plexus of SCD mice. Peripheral blood (5 µl) was drawn from the tail vein on d 1, 2, 3, 5, and 8 to detect biotinylated-labeled erythrocytes. Flow cytometric analysis was used to detect the fraction of biotinylated erythrocytes labeled with streptavidin-conjugated fluorochrome and Ter-119 to identify erythrocytes, as previously described (22).

Morphology study of erythrocytes

To determine the morphology of erythrocytes, blood smears were made using 1% glutaraldehyde fixed tail blood from SCD transgenic (Tg) mice. Blood smears were stained with WG16 (500-ml kit; MilliporeSigma). Blood smears stained by these procedures were observed using the ×100 oil immersion objective of a BX60 microscope (Olympus, Tokyo, Japan). Areas in which erythrocytes did not overlap were randomly picked, ≥10 fields were observed, and 1000 erythrocytes, including sickle cells, were counted. The percentages of sickle cells among the erythrocytes were calculated (23).

Mouse tissue isolation and histologic analysis

Mouse tissues were harvested and fixed in 10% phosphate-buffered formalin overnight. Fixed tissues were rinsed in PBS, dehydrated through graded ethanol washes, and embedded in paraffin. Tissue sections (5 µm) were stained with hematoxylin and eosin (H&E), according to the manufacturer’s instructions (Shardon Lipshaw, Pittsburgh, PA, USA). Semiquantification of histologic changes were analyzed in various tissues to identify vascular congestion in lung and necrosis in spleen, liver, and kidney, as previously described (26, 27); 10–15 fields/H&E-stained tissue were examined at ×20 magnification. Bright-pink regions in each tissue section were selected and quantified using Image Pro Plus 4.0 software (Media Cybernetics, Rockville, MD, USA). Necrotic regions in the liver and renal cortex were marked with the magical pen tool in Photoshop software (Adobe Systems, San Jose, CA, USA). Quantification of 10 digital images/mouse tissue at ×20 magnification was performed with Image Pro Plus software. Whole areas of each image were considered to be 100%. Percentages of pathologic areas for multiple tissues were recorded, and averages were obtained. Means ± sem were obtained from 3–5 mice/group.

Flow cytometry

Tissues were harvested into a tissue culture dish and teased apart into a single-cell suspension. Cells were collected in flow cytometry staining buffer, and the cell suspension was passed through a 40-μm strainer to eliminate clumps and debris and collected in a conical cell suspension tube. Before staining, cells were incubated with Fc block (BD Biosciences, Franklin Lakes, NJ, USA) per 10⁶ cells in 100 µl of staining buffer for 30 min at 4°C. The cells were then stained with antibodies (CD3, CD4, CD8, S1P1, and F4/F80) labeled with phycoerythrin, allophycocyanin, PerCP-CY5.5, APC-Cy7, and Pacific Blue. Events were acquired on a Gallios Flow Cytometer (Beckman Coulter, Sydney, NSW, Australia), and the data were analyzed with the Kaluza software (Beckman Coulter) (28).

Mouse bone marrow macrophage isolation and differentiation

Primary mouse macrophages were derived from mouse bone marrow, as described previously (29). Mice were euthanized by rapid cervical dislocation, skin was pulled from hind legs to expose femurs, and femurs were surgically removed. Bone marrow cells were flushed out of the bone cavity with 5 ml of macrophage culture medium [10% FBS, 20% L929 supplement, 5 ml glutamine, 5 ml penicillin/streptomycin in 350 ml of DMEM (MilliporeSigma)]. Cell suspension was centrifuged at 500 g for 10 min at room temperature; the supernatant was discarded, and the cells were resuspended in macrophage medium and counted. Cells (3 × 10⁶) in 7.5 ml of macrophage medium were added to a 100-mm Petri dish and incubated at 37°C, 5% CO₂ for 4 d. On d 4, an additional 5 ml of medium was added to the dish, and cells were cultured for 3–4 additional days. On d 7 or 8, the culture supernatant was discarded, and 5 ml of fresh macrophage medium was added to the dish. Cells were scraped and counted, and 2 × 10⁶ cells were added to each well of a 6-well plate. When the confluence of cells reached 80–90%, cells were washed twice with serum-free DMEM and treated with S1P and S1P receptor antagonists (final concentration: 1 μM each) or the Jak2 inhibitor AG490 (final concentration: 10 μM; Bio-Techne) for 24 h in serum-free DMEM. Cells were cultured at 37°C, 5% CO₂ for 7 d. On d 7, the medium was collected, filtered, and stored at −80°C.

Semiquantitative RT-PCR analysis

Total RNA was extracted from macrophages with TRIzol reagent (Thermo Fisher Scientific). RNase-free DNase (Thermo Fisher Scientific) was used to avoid genomic DNA contamination. Reverse transcription was performed using QuantiTect reverse transcription kit (Qiagen). Real-time PCR was performed on a Stratagene MxPro3000p (Agilent Technologies, Santa Clara, CA, USA) with SYBR Green Master Mix (Qiagen). Primers used for gene expression included forward IL-6, 5′-TCTGCAAGAGACTTCCATCCA-3′; reverse IL-6, 5′-CAGGTCTGTTGGGAGTGGTA-3′; forward S1PR1, 5′-GGGAGGGGACCCCAGCTCAG-3′; reverse S1PR1, 5′-CAGCCTCGCTCAAGCCGGAC-3′; glyceraldehyde-3-phosphate dehydrogenase forward, 5′-CAAGGTCATCCATGACAACTTTG-3′; and glyceraldehyde-3-phosphate dehydrogenase reverse, 5′-GGCCATCCACAGTCTTCTGG-3′. For data analysis, the ΔΔC_t (cycle threshold) method was used.

Statistical analysis

Results are expressed as means ± sem. All data were subjected to statistical analysis using 1-way ANOVA, followed by the Newman-Keuls post hoc test or the Student’s t test to determine the significance of any differences among groups. Statistical programs were run by Prism 5 statistical software (GraphPad Software, La Jolla, CA, USA). Statistical significance was set at P < 0.05.

RESULTS

FTY720 treatment has a potent anti-inflammatory and no antisickling effect in SCD mice

FTY720 is a functional antagonist for S1PR1 by inducing permanent endocytosis of S1PR1 and a U.S. Food and Drug Administration (FDA)–approved drug to treat multiple sclerosis as an immunosuppressant (30–34). To determine whether S1P-S1PR1 signaling contributes to chronic inflammation in SCD, we treated SCD mice with FTY720 (1 mg/kg/d in vehicle, i.p. injection) or saline (vehicle) for 8 wk. First, we monitored the expression of S1PR1 on LYs in lymph nodes and the thymus by flow cytometry in untreated WT mice. We found that the S1PR1 expression level was significantly higher in the CD3⁺CD4⁻CD8⁻ population than in any other populations (CD3⁺CD4⁺CD8⁺, CD3⁺CD4⁺CD8⁻, CD3⁺CD4⁻CD8⁺ in lymph nodes and thymus of WT mice compared with isotype controls) (Supplemental Fig. 1A, B). Thus, we chose to assess the therapeutic effects of FTY720 in mice by comparing S1PR1 expression in CD3⁺CD4⁻CD8⁻ population among saline- and FTY720-treated WT and SCD mice. After 8 wk of treatment, we found that S1PR1 expression in CD3⁺CD4⁻CD8⁻ cells was significantly reduced in FTY720-treated SCD mice compared with saline-treated SCD mice (Fig. 1A; P < 0.001), indicating that FTY720 treatment was effective at successfully reducing S1PR1 expression on LYs in SCD mice.

Figure 1.

Anti-inflammatory effects of FTY720 treatment in SCD mice. SCD mice were treated with saline or FTY720. After 8 wk of treatment, the following parameters were measured. A) S1PR1 expression on CD3⁺CD4⁻CD8⁻ (double-negative) T-cells from lymph nodes and thymus of WT and SCD mice treated with FTY720 or saline. B) FTY720 treatment reduced plasma cytokine levels in SCD mice. OD, optical density. C) FTY720 treatment reduced plasma IL-6 levels in SCD mice. D) FTY720 treatment reduced IL-6 expression levels in spleen, liver, lung, and kidney of SCD mice. E) No effect of FTY720 on red blood cell life span in SCD mice. Values shown represent means ± sem (n = 6–7/group). *P < 0.01, WT mice treated with saline compared with WT mice treated with FTY720; **P < 0.05, SCD mice treated with saline compared with WT mice treated with saline or SCD mice treated with FTY720 compared with SCD mice treated with saline, respectively; ^#P < 0.001, SCD mice treated with FTY720 compared with SCD mice treated with saline.

Next, we assessed the anti-inflammatory effect of FTY720 treatment in circulating immune cells. In CBC analysis, we discovered that FTY720 treatment significantly reduced white blood cells (WBCs), LYs, neutrophils (NEs), and monocytes (MOs) in the periphery (Table 1; P < 0.001). Additionally, using the cytokine ELISA array, we found that FTY720 treatment significantly decreased the increase in many inflammatory cytokines in the plasma, including IL-2, IL-12, IL-6, and IL-17 (Fig. 1B; P < 0.01). Circulating IL-6 was elevated in humans and mice with SCD and has long been suspected of contributing to chronic inflammation and tissue damage (22, 35). However, the molecular basis underlying elevated IL-6 in SCD remains unclear. Extending from the ELISA array results, we further confirmed that IL-6 was significantly increased in the plasma and in multiple tissues in SCD mice and that FTY720 treatment significantly reduced those increased levels in plasma and tissues in SCD mice (Fig. 1C, D). Thus, we provided pharmacologic evidence that FTY720 treatment reduced elevated plasma and tissue IL-6 levels in SCD mice, which implicates S1P signaling via its receptors as a contributor to elevated IL-6.

TABLE 1.

Hematologic parameters of SCD Tg mice treated with or without FTY720 for 8 wk and SCD/Il6^−/− mice

Mice (n=6)	RBCs (106/µl)	Hb (g/dl)	HCT (%)	MCV (fl)	RDW (%)	WBCs (103/µl)	Lys (103/µl)	NEs (103/µl)	MOs (103/µl)	EOs (103/µl)	BAs (103/µl)	Sickling (%)
SCD Tg + saline	5.09 ± 1.70	5.09 ± 1.70	18.055 ± 8.04	40.82 ± 2.59	34.05 ± 4.54	27.25 ± 1.31	20.43 ± 1.60	4.44 ± 0.4	2.26 ± 031	0.09 ± 0.02	0.04 ± 0.02	15.23 ± 3.2
SCD Tg + FTY720	5.74 ± 0.58	6.5 ± 1.94	23.10 ± 5.12	40 ± 5.93	30.47 ± 5.5.28	8.14 ± 0.69a	5.33 ± 0.79a	2.06 ± 0.36a	0.62 ± 0.13a	0.12 ± 0.04	0.01 ± 0.00	15.89 ± 2.6
SCD Tg/IL-6−/−	6.05 ± 0.81	6.36 ± 0.66	20.26 ± 1.69	48.2 ± 2.62	34.63 ± 4.30	15.33 ± 0.87a	11.49 ± 0.49a	2.41 ± 0.44a	1.20 ± 0.03a	0.14 ± 0.02	0.03 ± 0.00	16.23 ± 1.78

BA, basophil; EO, eosinophil; HCT, hematocrit; MCV, mean corpuscular volume; RBC, red blood cell; red blood cell distribution width.

^aP < 0.05 vs. SCD mice treated with saline.

Consistent with early in vitro studies showing that intracellular S1P contributes to sickling, independent of its receptors (22), we demonstrated that FTY720 treatment had no significant effect on sickling in SCD mice (Table 1). Supporting that finding, we observed no difference in red blood cell life span between SCD mice treated with saline and those treated with FTY720 (Fig. 1E). Altogether, we demonstrated that S1P signaling via S1PR1 contributes to chronic inflammation in SCD mice.

FTY720 treatment significantly reduces multitissue damage and improves kidney function in SCD mice

Next, we conducted a histologic analysis to determine whether FTY720-mediated anti-inflammatory effects have a role in reducing tissue damage in SCD mice. Specifically, microvascular congestion and hemolysis in the lung and microinfarction and cysts in the renal cortex were evident and are pointed out with arrowheads in (Fig. 2A) in SCD mice treated with saline. Moreover, necrotic regions were identified and indicated by arrowheads in liver and spleen sections in saline-treated SCD mice (Fig. 2A). Furthermore, vascular congestion and necrosis in lung, liver, spleen, and kidney were improved by FTY720 treatment in SCD mice (Fig. 2A). None of these pathologic changes were observed in saline-treated WT mice. Semiquantification revealed that the histologic changes in SCD mice treated with FTY720 were significantly improved compared with saline-treated SCD mice (Fig. 2B–E; P < 0.05). Consistent with the improved histologic analysis in the renal medulla region of the kidney, FTY720-treated SCD mice showed a reduction in proteinuria (Fig. 2F; P < 0.05), which indicated improved renal function. Altogether, FTY720 treatment reduced multitissue damage and ameliorated kidney dysfunction in SCD mice.

Figure 2.

FTY720 treatment or genetic deletion of Il6 reduces multitissue damage and kidney dysfunction in SCD mice. A) H&E staining of lung, spleen, liver, and kidney. Arrowheads denote representative vascular congestion in lung and to necrosis in spleen, liver, and kidney. Scale bars, 200 μm. B–E) Semiquantification of congestion or necrosis in lung, spleen, liver, and renal medulla. F) Ratio of microalbumin to creatinine in urine. G) Plasma IL-6 was abolished in SCD/Il6^−/− mice. *P < 0.01, SCD mice compared with WT mice; **P < 0.05, SCD mice treated with FTY720 compared with SCD mice treated with saline or SCD mice compared with SCD/Il6^−/− mice. ND, not determined.

Genetic deletion of Il6 reduces inflammation, multitissue damage, and kidney dysfunction in SCD mice without an effect on sickling

Because FTY720 was effective in reducing increased IL-6 levels in circulation and in the peripheral tissues (Fig. 1C–E), it is possible that elevated IL-6 is a key proinflammatory cytokine underlying systemic and local inflammation and contributing to tissue damage and dysfunction in SCD. To address that possibility, we generated SCD mice with global genetic deletion of the Il6 gene (See Supplemental Fig. 2). First, we confirmed that Il6 was successfully deleted by showing a complete elimination of circulating IL-6 in SCD/Il6^−/− mice, whereas IL-6 levels were still significantly elevated in SCD mice without the Il6 deletion (Fig. 2G; P < 0.05). Similar to FTY720 treatment, genetic deletion of Il6 had no effect on sickling (Table 1). However, CBC analysis showed that total WBCs, LYs, NEs, and MOs were significantly reduced in SCD/Il6^−/− mice compared with SCD mice (Table 1; P < 0.05), indicating that IL-6 has an important role in chronic inflammation in SCD mice.

Next, we compared tissue damage and renal dysfunction between SCD mice and SCD/Il6^−/− mice. Consistent with FTY720 treatment, histologic studies and semiquantification analysis revealed that multiple instances of tissue damage, including congestion and hemolysis in lungs and necrosis in the spleen, liver, and kidney, were significantly reduced in SCD/Il6^−/− mice compared with SCD mice (Fig. 2A–E). Moreover, genetic deletion of IL-6 in SCD mice also significantly attenuated proteinuria in SCD mice (Fig. 2F). Thus, we demonstrated that IL-6 is a key cytokine contributing to inflammation and multiple tissue damage in SCD mice.

FTY720 treatment and genetic deletion of Il6 attenuates elevated S1PR1 expression in tissue macrophages in SCD mice

Given the critical role of reciprocal regulation of IL-6 signaling cascade with S1P-S1PR1 in sustained inflammation in cancers and autoimmune diseases (5, 36) and in view of our findings that FTY720 treatment and genetic deletion of Il6 reduced chronic inflammation and tissue damage in multiple tissues, we hypothesized that excessive S1P-S1PR1 signaling mediated elevated IL-6 production reciprocally, inducing S1PR1 gene expression in tissue macrophages, was a potential mechanism underlying S1PR1 activation-promoted chronic inflammation and tissue damage in SCD mice. To test that hypothesis, we used flow cytometry to quantify expression of S1PR1 in F4/80-positive cells (macrophage marker) in several tissues isolated from WT and SCD mice treated with FTY720 or saline. First, we discovered that the expression of S1PR1 in F4/80⁺ populations of the spleen and kidneys of SCD mice was significantly induced compared with control mice (Fig. 3A, B). Moreover, we observed a significant reduction of expression of S1PR1 in F4/80⁺ populations of spleen and kidney in FTY720-treated mice compared with saline-treated SCD mice (Fig. 3A, B). Thus, we demonstrated that increased S1P is capable of inducing tissue macrophage S1PR1 expression in SCD mice.

Figure 3.

FTY720 treatment and genetic deletion of Il6 reduce elevated S1PR1 expression in tissue macrophages of SCD mice. A, B) Expression of S1PR1 in F4/80⁺ macrophages of spleen (A) and kidney (B) of WT and SCD mice treated with saline or FTY720. Data are expressed as means ± sem (n = 6–8/group).*P < 0.05, WT mice treated with FTY720 vs. saline-treated WT mice; **P < 0.05, SCD mice treated with saline vs. saline-treated WT mice; ^#P < 0.05, FTY720-treated SCD mice vs. saline treated SCD mice. C, D) Expression of S1PR1 in F4/80⁺ macrophages of spleen (C) and kidney (D) of control, SCD, and SCD/Il6^−/− mice. Data are expressed as means ± sem (n = 6–8/group). *P < 0.05, SCD mice compared with WT control mice; **P < 0.05, SCD mice compared with SCD/Il6^−/− mice.

Next, we assessed whether S1PR1-activation–mediated, elevated IL-6 was a key cytokine, capable of reciprocally promoting S1PR1 expression in tissue macrophages of SCD mice. Similar to FTY720 treatment, we discovered that the elevated expression of S1PR1 in F4/80⁺ cells of the spleen and kidney was significantly reduced in SCD/Il6 ^−/− mice compared with SCD mice (Fig. 3C, D; P < 0.05), indicating that elevated IL-6 underlies up-regulation of S1PR1 expression in tissue macrophages in SCD. Taken together, we provide both pharmacologic and genetic evidence that IL-6 is a critical cytokine induced by elevated, S1P-mediated S1PR1expression in macrophages and that elevated IL-6 reciprocally induces S1PR1 expression in macrophages of multiple tissues in SCD mice.

S1PR1 signaling induces IL-6 expression and IL-6 reciprocally promotes S1PR1 expression in primary mouse macrophages in a JAK2-dependent manner

It is difficult to assess the reciprocal regulation of elevated S1P-S1PR1 and IL-6/JAK2 in macrophages in an intact animal. To address that important question, we isolated bone marrow cells from C57BL/6 (WT) or Il6^−/− mice, followed by induction and selection of macrophages from bone marrow cells in macrophage medium (for details, see Materials and Methods). First, to determine whether S1P directly induces IL-6 gene expression via S1PR1 activation in macrophages, we stimulated WT mouse primary macrophages with S1P with DMSO or S1PR antagonists specific for S1PR1 (FTY720), S1PR1 (W146), S1PR2 (JTE013), S1PR3 (TY52156), and S1PR4 (CYM50358), at 1 µM. Similar to in vivo studies, we found that S1P induced IL6 gene expression and that this induction was significantly reduced by treatment of FTY720 or W146 (Fig. 4A). However, none of the other S1P receptor antagonists had an inhibitory effect on S1P-induced Il6 gene expression in primary cultured mouse macrophages (Fig. 4A). We also found that IL-6 protein levels in the culture medium were significantly induced by S1P treatment and that this induction was specifically blocked by FTY720 or W146, but not by other S1P receptor antagonists in primary, cultured mouse macrophages (Fig. 4B). Thus, we provide direct evidence that elevated S1P signaling via S1PR1 activation underlying increased Il6 gene expression in mouse primary macrophages.

Figure 4.

S1P signaling via S1PR1 directly induces Il6 gene expression and elevated Il6 reciprocally up-regulates S1PR1 gene expression in a JAK2-dependent manner in primary, cultured mouse macrophages. Mouse bone marrow from WT and Il6^−/− mice were isolated and differentiated to macrophages following with treatment of S1P in the presence or absence of S1P receptor specific antagonists or JAK2 inhibitor. A, B) Il6 gene expression (A) and protein level (B) in the medium of cultured WT mouse primary macrophages. C) S1pr1 mRNA levels in WT and Il6^−/− cultured primary macrophages. D–F) Il6 gene expression (D); secreted IL-6 protein levels (E); and S1pr1 gene expression (F) in cultured WT mouse primary macrophages with treatment of DMSO, S1P, or AG490 (a JAK2 inhibitor). Data are expressed as means ± sem (n = 6/group). *P < 0.01, S1P treated group compared with DMSO-treated group; **P < 0.05, S1P-treated group compared with FTY720-treated group, S1P-treated group compared with W146-treated group, and S1P-treated group compared with AG490-treated group. G) Working model of elevated S1P with dual consequences in SCD: 1) it functions intracellularly within erythrocytes to induce sickling (22); and 2) it functions extracellularly via S1PR1 signaling in macrophages to induce chronic inflammation and tissue damage. Elevated plasma S1P activating via S1PR1 on macrophages up-regulates Il6 gene expression and production in a JAK2-dependent manner that reciprocally induce S1pr1 gene expression. Treatment with FTY720 inhibits S1PR1 activation, or genetic deletion of Il6 successfully reduced sustained inflammation and tissue damage in SCD mice.

Next, to determine whether the S1P mediated induction of IL-6 was essential for S1P-induced S1pr1 gene expression in macrophages, we treated primary macrophage cells from WT and Il6^−/− mice with DMSO or S1P in the presence or absence of FTY720 or W146. S1pr1 gene expression was measured by semiquantitative RT-PCR of S1pr1 mRNA. Consistent with our in vivo findings, S1P treatment significantly induced S1pr1 mRNA in cultured primary macrophages from WT mice compared with DMSO-treated cells and that the induction was significantly attenuated in cultured, primary Il6^−/− mouse macrophages (Fig. 4C). Moreover, S1P-induced S1PR1 mRNA was attenuated to similar levels by the presence of FTY720 or W146 (Fig. 4C). Thus, we provide both genetic and pharmacologic evidence that elevated, S1P-induced IL-6 production promotes S1PR1 gene expression in macrophages.

JAK2 is a signaling molecule functioning downstream of S1PR1 activation and underlies up-regulation of S1pr1 and Il6 gene expression mediated by elevated S1P in tumors (36). However, the importance of JAK2 for reciprocal regulation of S1PR1 and IL-6 in macrophages remains undetermined. To test that possibility, we treated primary mouse macrophages with DMSO or S1P in the presence or absence of the specific JAK2 inhibitor, AG490. After 24 h, IL-6 and S1PR1 mRNA levels were determined, and the level of IL-6 in the culture medium was measured. The results show that S1P treatment induced S1pr1 and Il6 gene expression and IL-6 protein levels in the medium and that those increases were significantly reduced by AG490 (Fig. 4D–F). Altogether, our studies revealed that the S1P-induced up-regulation of S1PR1 and IL-6 signaling network in mouse macrophages is dependent on JAK2.

DISCUSSION

SCD is associated with prolonged, local and systemic inflammation, and persistently elevated inflammation has been long speculated as being detrimental in tissue damage and disease progression (20). However, specific factors and signaling pathways underlying sustained inflammation and tissue damage in SCD remain elusive. Of note, recent studies showed that SPHK1 activity was significantly induced in human and mouse SCD erythrocytes and that increased S1P, mediated by elevated erythrocyte SPHK1, contributes to sickling, independent of its receptors (22). However, the functional role of elevated, circulating S1P in prolonged inflammation and tissue damage in SCD remained unknown before this study. To address that important question, we conducted preclinical studies treating SCD mice with FTY720 (fingolimod; trade name, Gilenya; Novartis, Basel, Switzerland), a FDA-approved drug for the treatment of multiple sclerosis in humans (31). Here, we demonstrated that interfering with S1PR1 receptor activation by FTY720 reduced systemic and local tissue inflammation and tissue damage in SCD mice. Additionally, we provided genetic evidence that S1PR1-mediated, elevated IL-6 is essential for systemic inflammation and damage in multiple tissues in SCD. This finding led us to further discover that both FTY720 treatment and genetic deletion of Il6 reduced the S1PR1 levels in the tissue macrophages of SCD mice. Mechanistically, we showed that elevated S1P, which specifically activates S1PR1, underlies increased Il6 gene expression and that elevated Il6 reciprocally up-regulates S1pr1 gene expression in a JAK2-dependent manner in primary mouse macrophages. Overall, we provide both pharmacologic and genetic evidence that S1P-mediated S1PR1 activation is a key mechanism underlying chronic inflammation and tissue damage by reciprocally up-regulating S1pr1 and Il6 gene expression in macrophages, and we suggest that these signaling pathways are novel therapeutic targets for disease management (Fig. 4G).

FTY720/fingolimod is phosphorylated to form fingolimod-phosphate, which activates S1PR1, predominantly in immune cells; induces S1PR1 down-regulation; and in turn, prevents S1PR1-mediated inflammation (30). Consistently, we demonstrated FTY720 treatment significantly reduced inflammation systemically and locally and ameliorated damage to multiple tissues and kidney dysfunction in SCD mice. Circulating IL-6 is elevated in humans and mice with SCD and has been long suspected of contributing to chronic inflammation and tissue damage (22, 35). However, the molecular basis underlying its elevation and its specific role in SCD pathophysiology remain unclear. Here, we showed that plasma IL-6 was increased the most among all cytokines we measured in SCD mice. We further revealed that IL-6 levels in multiple tissues are elevated in SCD mice. Moreover, we found that down-regulation of S1PR1 by FTY720 reduced systemic and local tissue levels of IL-6 in SCD mice, implicating S1P signaling via S1PR1 as underlying elevated IL-6 in SCD mice. Extending from these findings, we demonstrated, for the first time to our knowledge, that genetic deletion of Il6 in SCD mice significantly reduced total circulating WBCs, LYs, MOs, and damage to multiple tissues. Although both FTY720 treatment and genetic deletion of Il6 has substantial anti-inflammatory and anti–tissue-damage effects in SCD mice, they do not have anti-sickling effects in SCD mice. Thus, our studies indicate that elevated IL-6 is a key cytokine functioning downstream of S1PR1 activation, underlying S1P-mediated chronic inflammation and tissue damage in SCD mice, independent of intracellular S1P-mediated sickling.

Because macrophages are the major immune cells in multiple tissues of SCD mice, macrophages are likely the key cell types for S1PR1–IL-6–induced tissue damage. Supporting that possibility, we demonstrated that FTY720 treatment and genetic deletion of Il6 substantially reduced S1PR1 expression in tissue macrophages in SCD mice. Thus, we have demonstrated that circulating S1P signaling via its receptors, most likely S1PR1, underlies the systemic and local tissue elevation of IL-6 and that elevated IL-6 is detrimental for SCD by promoting systemic inflammation, macrophages in multiple tissues and, in turn, inducing tissue damage in SCD mice. Early studies showed that S1P links persistent JAK2/p-STAT3 activation, chronic intestine inflammation, and development of colitis-related cancers (5). Moreover, S1P contributes to persistent JAK2 activation and sustained elevation of IL-6 by inducing S1PR1 in stromal cells of tumors (36). However, S1P-mediated elevation of IL-6 production leading to chronic inflammation and tissue damage by inducing S1PR1 on macrophages in SCD has not, to our knowledge, been previously reported. Using SCD/Il6^−/− mice, we found that genetic deletion of Il6 in SCD mice reduced S1pr1 gene expression in the macrophages of multiple tissues and reduced damage in multiple tissues. Additionally, we validated our in vivo findings in cultured, primary mouse macrophages and demonstrated that S1P directly induced Il6 gene expression and protein levels via activation of S1PR1 in a JAK2-dependent manner; and S1P-mediated elevation of IL-6 and subsequent activation of JAK2 underlies increased S1pr1 gene expression. As such, S1P-S1PR1-IL6-JAK2-S1PR1 functions as a malicious, positive feedback loop in macrophages to sustain chronic inflammation and multiple tissue damage in SCD. Thus, reducing S1P production by inhibiting SPHK1 and by interfering with S1PR1 activation and IL-6 signaling, it should be possible to reduce chronic inflammation, tissue damage, and disease progression in SCD mice.

Because of its strong, proinflammatory properties and its immune-stimulatory activities, excessive S1P is associated with a number of pathologic conditions (1). In particular, S1P contributes to pathophysiology associated with a number of autoimmune diseases, including multiple sclerosis, rheumatoid arthritis, Crohn’s disease, and chronic arthritis (37–39). The use of FTY720 to inhibit S1PR1 receptor activation has become an important therapeutic strategy for the treatment of multiple sclerosis (31). Because FTY720 is an FDA-approved drug, our preclinical studies set up a strong foundation for future clinical trials to treat patients with SCD with FTY720 to reduce chronic inflammation and slow disease progression. Moreover, neutralizing antibodies for IL-6 have also been used to successfully treat multiple immunologic disorders (40, 41). Our findings reveal that S1P-mediated S1PR1 activation and the induction of IL-6 in macrophages are critical to sustaining inflammation and tissue damage in SCD. Our results highlight the therapeutic potential of interfering with the S1P-S1PR1-IL6 signaling cascade for the treatment of SCD.

Supplementary Material

This article includes supplemental data. Please visit http://www.fasebj.org to obtain this information.

Glossary

CBC

complete blood cell count

H&E

hematoxylin and eosin

Hb

hemoglobin

LY

lymphocyte

MO

monocyte

NE

neutrophil

SCD

sickle cell disease

S1P

sphingosine-1-phosphate

SPHK1/2

sphingosine kinase 1/2

S1PR

sphingosine-1-phosphate receptor

S1PR1

sphingosine-1-phosphate receptor 1

Tg

transgenic

WBC

white blood cell

WT

wild type

AUTHOR CONTRIBUTIONS

Studies were designed by S. Zhao, M. G. Adebiyi, and Y. Xia; pharmacological and genetic studies were performed by S. Zhao, M. G. Adebiyi, and Y. Zhang; sample collection and flow cytometric analyses were performed by S. Zhao and M. G. Adebiyi; Y. Zhang conducted histologic analyses and macrophage culture and real-time PCR experiments; J. P. Couturier and D. E. Lewis guided flow cytometric experiments and provided advice on data analysis; X. Fan and H. Zhang helped with data analyses; R. E. Kellems helped edit the manuscript; S. Zhao and M. G. Adebiyi wrote the draft of the manuscript; Y. Xia supervised the study and worked closely with M. G. Adebiyi to revise the manuscript; and all authors contributed to manuscript review.